The goal of this proposal is to relate the structural effects of amino acid substitutions to protein evolution. Two Drosophila reproductive proteins and a cytochrome P450 will be altered to ask specific questions about adaptive evolution. Specific aim 1 will characterize the level of constraint that structural stability has on protein divergence. The neutral expectation of no constraint may be too conservative for the identification of positive selection, because there are still constraints on structure during adaptive evolution. This aim will use experimental and computational approaches to estimate a structurally neutral rate of evolution, which should be less than the non-constrained rate. The stability effects of random substitutions to the study proteins will be measured using denaturing techniques. A complementary approach will use computational tools to predict structural effects. A better understanding of structurally neutral evolution can aid in the prediction of disease-causing mutations and improve the power to detect episodes of adaptive evolution. Specific aim 2 will use the structural effects of historical substitutions to test hypotheses about positive selection, compensatory mutations, and the evolution of protein interactions. Ancestral protein sequences will be inferred and expressed so the effects of past substitutions can be measured. The stability effects of these substitutions will be used to determine which sequence of fixations could have occurred. These data will also be used to test whether selection for compensatory mutations could be responsible for a substantial proportion of positively selected substitutions. One study protein, ovulin, forms a homodimer. The effects of dimerization on this self-interaction will be used to test if dimerization gives ovulin extraordinary freedom to explore new sequences. Specific aim 3 will develop a method to infer whether two proteins have co-evolved. Co-evolution is implicated as a driving force behind positive selection. This aim will improve the statistical rigor of a phylogenetic technique designed by the applicant. It will be used to test the hypothesis of coevolution between the two reproductive proteins and candidate interactors. Co-evolution could explain their structural patterns of divergence. By interpreting the functional effects of substitutions we can infer how new phenotypes arose from adaptive change. PUBLIC HEALTH RELEVANCE Proteins form a large part of the functional apparatus of human cells and tissues. By understanding how their structures have evolved over time, we can make better predictions about their robustness to damaging mutations and then identify cases when they experience pressure to alter their functions.